Computational Investigation of Pathogen Evolution by Rachel Sealfon Submitted to the Department of Electrical Engineering and Computer Science - in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Computer Science MASSACES INSTTUTE OF TECHNOLOLGY at the JUL 07 2015 MASSACHUSETTS INSTITUTE OF TECHNOLOGY LIBRARIES June 2015 Massachusetts Institute of Technology 2015. All rights reserved. Signature redacted A uthor . .......................... Department of Electrical Engineering and Computer Science Z57 /May 18, 2015 ignature redacted Certified by.. / Pardis C. Sabeti Associate Professor /Z '/7 Signature redacted Thesis Supervisor Certified by.. ................... Manolis Kellis Professor Thesis Supervisor Signature redacted Accepted by ..... I// 1/0( Leslie A. Kolodziejski Chairman, Department Committee on Graduate Theses 2 Computational Investigation of Pathogen Evolution by Rachel Sealfon Submitted to the Department of Electrical Engineering and Computer Science on May 18, 2015, in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Computer Science Abstract Pathogen genomes, especially those of viruses, often change rapidly. Changes in pathogen genomes may have important functional implications, for example by al- tering adaptation to the host or conferring drug resistance. Accumulated genomic changes, many of which are functionally neutral, also serve as markers that can eluci- date transmission dynamics or reveal how long a pathogen has been present in a given environment. Moreover, systematically probing portions of the pathogen genome that are changing more or less rapidly than expected can provide important clues about the function of these regions. In this thesis, I (1) examine changes in the Vibrio cholerae genome shortly after the introduction of the pathogen to Hispaniola to gain insight into genomic change and functional evolution during an epidemic. I then (2) use changes in the Lassa genome to estimate the time that the pathogen has been circulating in Nigeria and in Sierra Leone, and to pinpoint sites that have recurrent, independent mutations that may be markers for lineage-specific selection. I (3) develop a method to identify regions of overlapping function in viral genomes, and apply the approach to a wide range of viral genomes. Finally, I (4) use changes in the genome of Ebola virus to elucidate the virus' origin, evolution, and transmission dynamics at the start of the outbreak in Sierra Leone. Thesis Supervisor: Pardis C. Sabeti Title: Associate Professor Thesis Supervisor: Manolis Kellis Title: Professor 3 4 Acknowledgments I am deeply grateful to my advisors Pardis and Manolis for their guidance and men- torship throughout my graduate years. I thank Pardis for welcoming me into the world of viral genomics, for creating an exceptionally warm and supportive lab envi- ronment, and for teaching me so much. I thank Manolis for being so extraordinarily supportive and encouraging me to find my own path in computational biology. I also appreciate the guidance of Constantinos Daskalakis, my other thesis committee member. My thanks to all my labmates in both the Sabeti and Kellis labs and to my other collaborators. Thanks especially to Kristian Andersen, Stephen Gire, Irwin Jungreis, Mike Lin, Daniel Park, and Maxim Wolf for close collaborations. Thanks to my friends. Thanks to my family for their love and support. I want to acknowledge the in- spiration of my grandfather, Boaz Gelernter, who escaped the Holocaust and became the first PhD (and an engineer) in my family, and of my endlessly loving and de- voted grandmother Gertrude Zinman Gelernter, who sadly passed away while I was completing this work. 5 6 Contents 1 Introduction 11 1.1 Human pathogen diversity ..... ........ 13 1.2 Pathogen evolution ..... ........ .... .... ..... 14 1.3 Pathogen sequencing ... ...... ...... .. ....... 15 1.3.1 Sequencing technologies ....... ... ......... 16 1.3.2 Assembly and alignment .. ... ... .. ......... 17 1.3.3 Challenges in sequencing pathogen genomes ......... 18 1.4 Phylogenies .. .... .... .... .... ... 18 1.4.1 Phylogenetic methods .. .... .... 18 1.4.2 Pathogen phylogenies ..... ..... .. 19 1.4.3 Inferring transmission .... ..... .. 19 1.5 Nucleotide and codon substitution models . ... 20 1.5.1 Nucleotide substitution models .. .... 20 1.5.2 Codon substitution models .... ..... 20 1.5.3 Model comparison and fitting .. ...... 21 1.6 Open data sharing for accelerating collaborative study of infectious outbreaks .... ........ ........ .. 21 1.7 Thesis contributions ..... ......... .. 22 2 High depth, whole-genome sequencing of cholera isolates from Haiti and the Dominican Republic 27 2.1 C ontributions . ...... ...... ...... ..... ...... 27 2.2 B ackground .. ...... ...... ...... ..... ...... 27 7 2.3 M ethods .. ..... ..... ..... ...... .... ..... .. 29 2.3.1 V. cholerae samples .. ..... ..... .... ..... ... 29 2.3.2 Sample preparation/ isolation ... .... ... ... ... .. 30 2.3.3 Illumina-based whole genome sequencing .. ... .. ... .. 31 2.3.4 De novo assembly .. .. ... .. .. ... .. .. ... .. .. 32 2.3.5 Comparison of sequence variants across sequencing technologies 32 2.3.6 Identifying SNPs, insertions, deletions, and structural variation across isolates .. ... ... ... ... ... ... ... ... 32 2.3.7 Constructing a phylogeny .. ... .... ... ... ... .. 36 2.4 Results and Discussion .. .... ... .... ... .... .... .. 36 2.4.1 Sequencing seven V. cholerae isolates at high depth of coverage 36 2.4.2 Effect of depth of coverage on genome assembly and single- nucleotide polymorphism (SNP) calling .. .... ... .... 36 2.4.3 Comparison of sequence variants, insertions, and deletions iden- tified using multiple sequencing approaches .. ... ... ... 38 2.4.4 Identifying SNPs, insertions, deletions, and structural variation across isolates .. ... .... ... ... ... ... ... ... 40 2.4.5 Analysis of Dominican Republic and Haitian isolates ... .. 41 2.4.6 Functional annotation of variants in Haitian and Dominican Republic cholera strains .. .. ... ... .. ... .. ... 45 2.5 C onclusions . ... ... ... ... .... ... ... ... ... ... 46 3 Molecular dating and evolutionary dynamics of Lassa virus 49 3.1 Contributions ... ... .... .... ... .... ... .... ... 49 3.2 Background . ... .. ... .. ... .. ... .. ... .. ... .. 49 3.3 M ethods . .. ... .. ... .. ... .. ... .. ... .. ... .. 51 3.3.1 Description of virus sequences that were analyzed ... .. .. 51 3.3.2 Molecular dating ... .. ... ... .. ... .. ... .. .. 52 3.3.3 Detecting lineage-specific rate variation . .. ... .. ... .. 52 3.4 Results and Discussion . ... .. ... .. ... .. ... .. ... .. 53 8 3.4.1 Molecular dating estimates are robust across many evolutionary models and data subsets ..................... 53 3.4.2 Randomization testing and root-to-tip linear regression support the presence of temporal structure in the data . ..... ... 57 3.4.3 Molecular dating indicates that LASV is ancient in Nigeria and shares a recent common ancestor outside Nigeria ...... .. 59 3.4.4 Detecting sites evolving with lineage-specific rates of change in LASV ........ ............................... 60 3.5 C onclusions .... ........ ......... ........ ... 62 4 Finding regions of excess synonymous constraint in diverse viruses using a phylogenetic codon substitution model-based framework 65 4.1 Contributions .......... ............. ........ 65 4.2 Background .. ......... ........ ........ ..... 65 4.3 M ethods ... ........ ........ ......... ...... 68 4.4 Results and Discussion .......... ............. ... 70 4.4.1 Finding Regions of Excess Synonymous Constraint (FRESCo): a phylogenetic codon-model based approach for detecting re- gions with reduced synonymous variability ........... 70 4.4.2 FRESCo displays high specificity in recovering regions of excess synonymous constraint in simulated sequences ... ...... 71 4.4.3 FRESCo recovers regions of known excess synonymous con- straint in well-characterized viral genomes: Hepatitis B virus, West Nile Virus, and poliovirus ...... ........... 74 4.4.4 FRESCo identifies known and novel regions of excess synony- mous constraint in 30 virus genomes .. ............ 76 4.4.5 Pinpointing regions of excess synonymous constraint near the 5' and 3' terminal regions of rotavirus segments ........ 77 4.4.6 Identifying novel candidate overlapping elements in bluetongue v iru s ... ..... .... .... .... .... ..... ... 78 9 4.4.7 Identifying novel regions of excess synonymous constraint with conserved, stable predicted RNA structure ........ .. 80 4.4.8 Identifying novel regions of excess synonymous constraint in Ebola virus and Lassa virus .................. 83 4.5 C onclusions: ............................... 85 5 Computational analysis of Ebola virus origin and transmission dur- ing the 2014 West Africa Outbreak 87 5.1 Contributions ... .... .... .... .... .... ...... 87 5.2 Introduction ... .... .... .... .... .... .. 88 5.3 Materials and Methods . ..... ..... ..... ... 90 5.3.1 Sample collection and sequencing .... ..... 90 5.3.2 Demultiplexing of raw Illumina sequencing reads . 90 5.3.3 Assembly of full-length EBOV genomes ... ... 91 5.3.4 Multiple sequence alignments ..... ...... 91 5.3.5 SN P calling .... ...... ...... ..... 91 5.3.6 Phylogenetic tree construction .. ......... 92 5.3.7 Counting fixed and variable polymorphic positions for each out- break . ........ ...... ... ........ ..... 92 5.3.8 Intrahost variant calling and analysis ... .... ...... 93 5.4 Results and Discusssion .. .........